| 01:09.20 | *** join/#brlcad ownjsyygcdafzfol (~armin@dslb-092-074-254-134.092.074.pools.vodafone-ip.de) | |
| 01:22.15 | *** join/#brlcad infobot (ibot@rikers.org) | |
| 01:22.15 | *** topic/#brlcad is BRL-CAD and open source CAx discussion ! Also @ http://brlcad.zulipchat.com ! Logs @ http://infobot.rikers.org/%23brlcad/ | |
| 01:29.48 | *** join/#brlcad LordOfBikes (~armin@dslb-088-064-127-195.088.064.pools.vodafone-ip.de) | |
| 01:48.35 | *** join/#brlcad LordOfBikes (~armin@dslb-092-075-150-068.092.075.pools.vodafone-ip.de) | |
| 02:21.30 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) | |
| 02:40.50 | *** join/#brlcad LordOfBikes (~armin@dslb-092-075-150-068.092.075.pools.vodafone-ip.de) | |
| 03:56.06 | starseeker | cool: https://github.com/rib/gputop/ |
| 03:56.06 | gcibot | [ GitHub - rib/gputop: A GPU profiling tool ] |
| 04:35.43 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) | |
| 09:34.01 | *** join/#brlcad teepee (~teepee@unaffiliated/teepee) | |
| 10:50.26 | *** join/#brlcad teepee- (bc5c2133@gateway/web/freenode/ip.188.92.33.51) | |
| 11:13.17 | *** join/#brlcad merzo (~merzo@185.39.197.205) | |
| 13:05.54 | *** join/#brlcad yorik (~yorik@2804:431:f721:848e:290:f5ff:fedc:3bb2) | |
| 13:44.25 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) | |
| 14:29.12 | *** join/#brlcad merzo (~merzo@185.39.197.205) | |
| 14:43.40 | *** join/#brlcad witness (uid10044@gateway/web/irccloud.com/x-aubyoavidfzpbhfw) | |
| 14:46.27 | *** join/#brlcad DaRock (~Thunderbi@mail.unitedinsong.com.au) | |
| 16:47.49 | *** join/#brlcad infobot (ibot@rikers.org) | |
| 16:47.49 | *** topic/#brlcad is BRL-CAD and open source CAx discussion ! Also @ http://brlcad.zulipchat.com ! Logs @ http://infobot.rikers.org/%23brlcad/ | |
| 17:28.07 | ``Erik | neat, even talks to vulcan O.o |
| 17:30.25 | *** join/#brlcad Notify (~notify@104.225.5.10) | |
| 17:32.32 | Notify | 03BRL-CAD:starseeker * 70716 brlcad/trunk/src/nirt/main.cxx: Ah, that's right - we need to help bu_brlcad_root out explicitly. |
| 17:32.48 | Notify | 03BRL-CAD:starseeker * 70710 brlcad/trunk/regress/CMakeLists.txt: Add depends_list not just to overall regress target but to the specific regression target in question as well. |
| 17:32.53 | Notify | 03BRL-CAD:starseeker * 70713 brlcad/trunk/src/nirt/main.cxx: right, jumps don't like crossing initializations... |
| 17:33.19 | Notify | 03BRL-CAD:starseeker * 70712 brlcad/trunk/src/nirt/main.cxx: initialize |
| 17:34.45 | ``Erik | app world is lucrative and fun, notify choked, not particularly (but iOS dev is moving to swift, which has some very lispy facets... objc feels painfully unlispy and crude now) |
| 17:41.05 | Notify | 03BRL-CAD:starseeker * 70719 brlcad/trunk/src/libanalyze/nirt.cxx: don't report gaps with gap_los <= 0 |
| 17:41.10 | Notify | 03BRL-CAD:starseeker * 70720 brlcad/trunk/regress/nirt.sh: Ah, that's what the GAP bits were in the regression diff - zero length reporting. |
| 17:41.26 | Notify | 03BRL-CAD:starseeker * 70721 brlcad/trunk/src/libanalyze/nirt.cxx: start stubbing in nirt diff command |
| 17:41.28 | Notify | 03BRL-CAD:starseeker * 70722 brlcad/trunk/src/libanalyze/nirt.cxx: allow comments |
| 17:44.10 | Notify | 03BRL-CAD:brlcad * 70724 brlcad/trunk/src/libfb/if_ogl.c: explicitly include bio.h for snprintf |
| 17:45.15 | *** join/#brlcad DaRock (~Thunderbi@mail.unitedinsong.com.au) | |
| 17:45.39 | Notify | 03BRL-CAD:starseeker * 70725 brlcad/trunk/src/libanalyze/nirt.cxx: start experimenting with std::precision to get precise control of string<->double conversions |
| 17:45.47 | Notify | 03BRL-CAD:starseeker * 70726 brlcad/trunk/src/libanalyze/nirt.cxx: If we're at the init stage, can't count on an error callback being available - go with bu_log |
| 17:45.51 | Notify | 03BRL-CAD:starseeker * 70727 brlcad/trunk/src/libanalyze/nirt.cxx: read in hit |
| 17:46.02 | Notify | 03BRL-CAD:starseeker * 70728 brlcad/trunk/src/libanalyze/nirt.cxx: handle comments better |
| 17:46.41 | Notify | 03BRL-CAD:starseeker * 70729 brlcad/trunk/src/libanalyze/nirt.cxx: More input logic for diffing |
| 17:46.44 | Notify | 03BRL-CAD:starseeker * 70730 brlcad/trunk/src/libanalyze/nirt.cxx: stash a couple reference links |
| 17:49.57 | Notify | 03BRL-CAD:starseeker * 70731 brlcad/trunk/src/libanalyze/nirt.cxx: For diff cataloging, the one-shot nirt output container is inadequate. Start breaking it down into components. |
| 17:54.04 | Notify | 03BRL-CAD:brlcad * 70732 brlcad/trunk/misc/CMake/BRLCAD_CheckFunctions.cmake: fix a couple issues in the std_hdr_defs function. if a HAVE_ test failed (i.e., was unset), it was still defining it (and setting to an empty string). empty -DHAVE_..="" strings then didn't play well with the #if tests which should have been #ifdef tests. lastly, made a change to only calculate the cppflags once since it doesn't need to |
| 17:54.06 | Notify | recheck the template again afterevery symbol. |
| 17:54.08 | Notify | ... |
| 17:59.28 | Notify | 03BRL-CAD:brlcad * 70750 brlcad/trunk/src/nirt/main.cxx: needs popen/pclose and fstream decls for c90/posix.1 compilation |
| 17:59.30 | Notify | 03BRL-CAD:brlcad * 70749 brlcad/trunk/src/mged/mged.c: convert from librt timer to libbu timer. need to test more to make sure ranges are still appropriate. |
| 17:59.32 | Notify | 03BRL-CAD:brlcad * 70741 brlcad/trunk/src/libtermio/termio.c: O_NONBLOCK is apparently available in c90/posix.1 api |
| 17:59.34 | Notify | 03BRL-CAD:brlcad * 70748 (brlcad/trunk/src/conv/intaval/tgf-g.cpp brlcad/trunk/src/conv/k-g.cpp and 4 others): more fstream decls |
| 17:59.36 | Notify | 03BRL-CAD:brlcad * 70747 brlcad/trunk/src/adrt/tienet.c: need decls |
| 17:59.38 | Notify | 03BRL-CAD:brlcad * 70739 brlcad/trunk/src/librt/CMakeLists.txt: obsolete sysv timer |
| 18:00.03 | Notify | 03BRL-CAD:brlcad * 70751 brlcad/trunk/src/libged/red.c: REG_STARTEND isn't POSIX.2 so try to rework the code. of the four places using REG_STARTEND in regexec(), it doesn't look like three actually need it since they scanned from so=0. however, one did so we try a rewrite which increments the string offset. |
| 18:00.09 | Notify | 03BRL-CAD:brlcad * 70752 brlcad/trunk/src/librt/timer42.c: don't have anything but ru_utime and ru_stime if we're compiling posix. |
| 18:00.12 | Notify | 03BRL-CAD:brlcad * 70753 brlcad/trunk/CMakeLists.txt: to specify c90/posix era API and get the right set of functiondeclarations and other other symbols from headers, we must define both_POSIX_C_SOURCE and _XOPEN_SOURCE. they configure different standards.the intent is after confirming 1990 standard's support, we'll jumpforward to at least 2013 where we will be safe to assume c11/c++11.added a comment for the |
| 18:00.14 | Notify | nominal timeframe targets that would makesense to target and what corresponding *_SOURCE values were current atthat time. update cmake logic accordingly to use a set of them(currently using the 1990 and 2000 timeframe sets for debug andrelease respectively). the one stray _XOPEN_SOURCE=500 test forsys/ipc.h, now removed/incorporated, was a critical culprit causinginconsistency between cmake function testing and flags |
| 18:00.16 | Notify | used duringcompilation.pulled the sys/sysctl.h header test out as it's a "broken" header onbsd when compiling in posix mode -- have to define u_int, u_short,u_long, u_char types. also added tests for fseeko (implying ftello),gettimeofday, and time functions.... |
| 18:00.18 | Notify | 03BRL-CAD:brlcad * 70754 brlcad/trunk/CMakeLists.txt: clean up so popen is back in the main list |
| 18:00.34 | Notify | 03BRL-CAD:starseeker * 70755 brlcad/trunk/doc/docbook/system/man5/CMakeLists.txt: Start working on a man page specific to NIRT's formatting functionality. This is borderline between a file format and command documentation... we are essentially using it as a file format by defining pre-packaged NIRT formatting files. Also, the topic is sufficiently complex/verbose that it warrants its own discussion and examples |
| 18:00.36 | Notify | rather than hiding it in a bigger NIRT man page. |
| 18:00.38 | Notify | ... |
| 18:00.47 | Notify | 03BRL-CAD:starseeker * 70756 brlcad/trunk/src/libanalyze/nirt.cxx: checkpoint |
| 18:01.20 | Notify | 03BRL-CAD:starseeker * 70757 brlcad/trunk/src/libanalyze/nirt.cxx: initialize out_old |
| 18:01.56 | Notify | 03BRL-CAD:brlcad * 70758 (brlcad/trunk/src/conv/bot_shell-vtk.c brlcad/trunk/src/conv/g-shell-rect.c and 31 others): static init is usually ever so slightly preferred since it gives the compiler a little more flexibility with instruction ordering. saves a few lines too. |
| 18:02.05 | Notify | 03BRL-CAD:brlcad * 70759 brlcad/trunk/CMakeLists.txt: pretty printing cleanup |
| 18:02.07 | Notify | 03BRL-CAD:brlcad * 70760 brlcad/trunk/src/libbu/file.c: c99 portability for S_ISLNK() |
| 18:02.20 | Notify | 03BRL-CAD:brlcad * 70761 brlcad/trunk/misc/CMake/BRLCAD_CheckFunctions.cmake: need to include sys/time.h for gettimeofday on BSD |
| 18:02.22 | Notify | 03BRL-CAD:brlcad * 70762 (brlcad/trunk/src/libfb/fbserv_obj.c brlcad/trunk/src/libfb/if_remote.c brlcad/trunk/src/libfb/if_tk.c): use the bnetwork.h header instead of directly including netinet/in.h |
| 18:02.24 | Notify | 03BRL-CAD:brlcad * 70763 brlcad/trunk/include/bnetwork.h: declare htonl/ntohl as needed since they are not included in c90/c99. |
| 18:02.26 | Notify | 03BRL-CAD:brlcad * 70764 brlcad/trunk/CMakeLists.txt: don't put API limitations on C++ compilation. C++ headers are nowhere near robust enough to have common posix symbols and types they rely on masked. test for htonl since it's not even c99. |
| 18:02.28 | Notify | 03BRL-CAD:brlcad * 70766 brlcad/trunk/src/librt/primitives/bot/gct_decimation/meshdecimationSSE2.c: use FMIN since fmin isn't part of c99. |
| 18:02.30 | Notify | 03BRL-CAD:brlcad * 70765 brlcad/trunk/src/librt/primitives/bot/gct_decimation/meshoptimizer.c: go double since pow() is everywhere and powf() came much later |
| 18:02.35 | Notify | 03BRL-CAD:brlcad * 70767 brlcad/trunk/CMakeLists.txt: wow, a lot more insideous debugging required to find a workaround for compiling c99 on Mac. turns out their system header logic drops the inline keyword (while still defining all the ctype functions) resulting in tons of duplicate symbols at link time. fix is to either not support c99 or this hack: define inline inline. the logic leaves it alone and we get |
| 18:02.37 | Notify | proper inlining. |
| 18:02.39 | Notify | ... |
| 18:02.41 | Notify | 03BRL-CAD:brlcad * 70768 brlcad/trunk/CMakeLists.txt: timestamp how recent the issue is a bit closer to reality |
| 18:03.03 | Notify | 03BRL-CAD:brlcad * 70769 (brlcad/trunk/src/conv/comgeom/region.c brlcad/trunk/src/conv/comgeom/solid.c): don't need these ctype decls, prevent inlining |
| 18:03.22 | Notify | 03BRL-CAD:brlcad * 70770 brlcad/trunk/misc/CMake/BRLCAD_Summary.cmake: the -std= flag is already in the cflags/cxxflags list, so skip the logic that appends them to the summary. |
| 18:03.28 | Notify | 03BRL-CAD:starseeker * 70771 brlcad/trunk/misc/CMake/BRLCAD_CheckFunctions.cmake: need netinet/in.h for htol declaration |
| 18:03.46 | Notify | 03BRL-CAD:starseeker * 70772 brlcad/trunk/src/libanalyze/nirt.cxx: Add some guards for report printing to avoid inactive keys getting used |
| 18:03.51 | Notify | 03BRL-CAD:starseeker * 70773 brlcad/trunk/src/libanalyze/nirt.cxx: add a note that this validation step should be pushed up to the fmt command... |
| 18:04.54 | Notify | 03BRL-CAD:starseeker * 70774 brlcad/trunk/src/libanalyze/nirt.cxx: checkpoint |
| 18:05.03 | Notify | 03BRL-CAD:starseeker * 70775 brlcad/trunk/src/libanalyze/nirt.cxx: checkpoint |
| 18:05.11 | Notify | 03BRL-CAD:brlcad * 70776 brlcad/trunk/src/libbu/backtrace.c: declare gettimeofday if needed |
| 18:05.13 | Notify | 03BRL-CAD:brlcad * 70777 brlcad/trunk/src/libbu/affinity.c: c90-mode hack for clockid_t, though still need it for cmake's package testing too |
| 18:06.31 | Notify | 03BRL-CAD:starseeker * 70778 brlcad/trunk/src/libanalyze/nirt.cxx: checkpoint |
| 18:07.22 | Notify | 03BRL-CAD:starseeker * 70779 brlcad/trunk/src/libanalyze/nirt.cxx: typo |
| 18:07.25 | Notify | 03BRL-CAD:brlcad * 70780 brlcad/trunk/CMakeLists.txt: disable march=native for now |
| 18:07.27 | Notify | 03BRL-CAD:brlcad * 70781 brlcad/trunk/include/analyze.h: match the implementation, size_t |
| 18:07.29 | Notify | 03BRL-CAD:brlcad * 70782 brlcad/trunk/include/CMakeLists.txt: brlcad_version.h must be installed because it's referenced in other headers |
| 18:07.32 | Notify | 03BRL-CAD:starseeker * 70783 brlcad/trunk/src/libanalyze/nirt.cxx: whoops - check for diff before trying anything |
| 18:07.34 | Notify | 03BRL-CAD:starseeker * 70784 brlcad/trunk/src/libanalyze/nirt.cxx: if we're going to shoot, need to make sure we're set up for it |
| 18:07.39 | Notify | 03BRL-CAD Wiki:Sean * 10492 /wiki/Google_Summer_of_Code/Project_Ideas: /* Web Development */ |
| 18:07.41 | Notify | 03BRL-CAD Wiki:Sean * 10493 /wiki/Google_Summer_of_Code/Project_Ideas: /* Geometry */ |
| 18:07.43 | Notify | 03BRL-CAD Wiki:Sean * 10495 /wiki/Google_Summer_of_Code/Project_Ideas: /* Infrastructure */ |
| 18:07.45 | Notify | 03BRL-CAD Wiki:Inderpreet * 10497 /wiki/Online_Geometry_Viewer_Back-end_GSOC_2018: Creating new page for gsoc 2018 backend OGV project |
| 18:07.47 | Notify | 03BRL-CAD Wiki:Inderpreet * 10498 /wiki/Google_Summer_of_Code/Project_Ideas: /* Web Development */ |
| 18:07.49 | Notify | 03BRL-CAD Wiki:Inderpreet * 10499 /wiki/Online_Geometry_Viewer_Back-end: Replaced content with "# Online Geometry Viewer Back-end for GSOC 2018" |
| 18:07.52 | Notify | 03BRL-CAD Wiki:Inderpreet * 10500 /wiki/Online_Geometry_Viewer_Back-end: |
| 18:07.54 | Notify | 03BRL-CAD Wiki:Inderpreet * 10501 /wiki/Online_Geometry_Viewer_Back-end: |
| 18:07.56 | Notify | 03BRL-CAD Wiki:Inderpreet * 10502 /wiki/Online_Geometry_Viewer_Back-end: |
| 18:07.58 | Notify | 03BRL-CAD Wiki:Inderpreet * 10503 /wiki/Google_Summer_of_Code/Project_Ideas: /* Web Development */ |
| 18:08.00 | Notify | 03BRL-CAD Wiki:Inderpreet * 10504 /wiki/Google_Summer_of_Code/Project_Ideas: /* Mentors */ |
| 18:08.02 | Notify | 03BRL-CAD Wiki:Inderpreet * 10505 /wiki/Mediawiki_3D_Geometry_Extension: Created page with " |
| 18:08.04 | Notify | 03BRL-CAD Wiki:45.248.165.181 * 10506 /wiki/Google_Summer_of_Code/Project_Ideas: /* Mentors */ |
| 18:08.06 | Notify | 03BRL-CAD Wiki:45.248.165.181 * 10507 /wiki/Google_Summer_of_Code/Project_Ideas: /* Web Development */ |
| 21:41.25 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) | |
| 21:57.41 | *** join/#brlcad ``Erik (~erik@pool-100-16-20-94.bltmmd.fios.verizon.net) | |
| 22:27.21 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) | |
| 23:06.53 | starseeker | growls at sourceforge |
| 23:26.13 | starseeker | brlcad: so for 70777 - do we need to add that bit to the CMake testing? |
| 23:27.43 | *** join/#brlcad Radicarian (~Radicaria@cpe-72-231-246-183.buffalo.res.rr.com) | |